Study Guide #3

What is a meta-analysis?

Meta-analysis is the systematic variance across time; analyzes results from a set of completed studies,... assesses the relationship between whatever variables are the focus of their analysis

standard normal distribution

A normal distribution with a mean of 0 and a standard deviation of 1.

What is a Z-Score? What does a Z-Score tell you?

A z-score is a standardized measure of the precise location of each X value within a distribution. It tells you the distance between a score and mean in standard deviation units.

outlier

An outlier is an individual value that is significantly different from other values of the same data.

central tendency

Central tendency is a statistical measurement that attempts to determine the single value usually located at the center of the data

Explain the concept of Least Squares (achieving the minimum SS in your data).

It decreases error by getting a minimum SS from the balance point which is the mean.

Which is best measure of central tendency, the mean, median, or mode? Why?

The best measure depends. Mean is preferred. Use median for 1. extreme scores/ skewed distributions 2. undetermined values 3. open-ended distributions. Use mode for 1. nominal scales 2. discrete variables and 3. describing shape.

Which of the measures of central tendency is most influenced by outliers in your data?

The mean is strongly influenced by outliers because it is the average of the scores and it is not resistant. The median is resistant.

Explain how you can start with a data set, and standardize this data set to any mean and standard deviation that you choose.

The original scores are transformed into z-scores. Z-scores are then turned into X-values so that the new mean and standard deviation are attained. raw score -> z-score -> standardized score Xnew = population mean desired + (standard deviation desired x Z

Sum of Squares? (SS)

The sum of squares is the sum of squared deviations about the mean.

When we compute the sample variance, why do we divide SS by n - 1 instead of n?

We divide it by n-1 because it rids the result of bias. Because it is a sample statistic it varies every time so the average of variance will never be truthful. Variance differs because of a varying mean. If we divide by n it is biased and systematically too small, but if we divide by n-1 it will equal the average of the population variance

Explain variance conceptually and computationally (make sure to use the term "deviations").

conceptually: variance is the average squared deviations about the mean computationally:

What are some factors that might contribute to error variance in your research?

- mistakes in recording data -coding data - most likely: factors that remain unidentified in a study (personal attributes or situational factors: temperature, mood)

Explain: total variance, systematic variance, and error variance.

1. total variance: systematic variance + error variance 2. systematic variance: variance in behavior explained by research variables 3. error variance: variance in behavior not explained by research variables

conceptually and computationally define the mean, median, or and mode

Conceptually: 1. mean: arithmetic average of the scores 2. median: score in the middle of the data 3. mode: most frequent score Computationally: 1. mean: 2. median: U (1/2) 3. mode:

We often use the term, "dispersion" to refer to variability in data. Why is this a particularly useful term when one is working to understand the concept of variance?

Dispersion is hot the data is spread out. This is useful because 1. variability describes the distribution; whether they cluster or are spread out 2. variability measures how an individual score represents the entire distribution

What is effect size?

Effect size measures how strongly variables in a study are related to one another, it is a statement of relationship between variables; assess the strength of the relationship

Describe how all research questions in psychology are questions about behavioral variability.

Psych questions are about behavioral variability because researchers are always studying a change in performance. These include the variability across situations, variables, and time. 1. research questions are about the causes and correlates of behavioral variability 2. researchers try to design studies that best explain the variability of a particular behavior 3. measures used in research attempt to capture numerically the variability in participants behavior 4. statistics are used to analyze the variability in our data

What is skewness, and what would be the relationship between the mean, median, and mode in a positively skewed distribution? ...in a negatively skewed distribution?

Skewedness is a statistical distribution in which the curve appears distorted to either the left or right; a measurement of asymmetry. Skewed to the left (negative) means the mean is less than the median and the median is less than the mode. Skewed right (positive) means that the mean is more than the median and the median is more than the mode.

What does it mean to standardize data? What are the benefits of standardizing data?

To standardize data, it means that a given object or measurement is interpreted or used in a consistent way across time, setting, and circumstance. The benefits of standardizing data are that scores are now comparable and it gives a better view of the population.